A model based on eye movement data and artificial neutral network for product styling evaluation

Abstract

The product styling evaluation method is usually subjective, and the lack of objectively quantitative analysis and data support. To make up these disadvantages and to obtain the objective assessment for product styling, a styling evaluation model based on eye-tracking technology and BP (back propagation) neural network was proposed, whose inputs were multiple eye movement data. This pilot study took the aircraft passenger seats as example. The experiment in this paper used RED desktop eye tracker to collect the eye movement data of 14 participants for evaluating eight aircraft passenger seats' styling, meanwhile, the subjective ratings were obtained with ratings scale via verbal self-report. A mathematical model for evaluating aircraft passenger seat styling was set up. The best number of the neurons in the hidden layer was determined by the training error. Finally, the model was verified. The result suggested that the model based on BP neural network could effectively calculate the subjective evaluation ratings on aircraft passenger seat styling, and this method provide objective support for evaluation. This model assists the designers and managers choose the optimal product design proposal and capture the user's preference.